Artificial Intelligence for Modeling and Simulation (AIMS) Lab

My previous research mainly focused on data-driven turbulence modeling by using Bayesian inference and machine learning techniques. When I was at Lawrence Berkeley National Laboratory, I also explored generative learning techniques (e.g. generative adversarial networks) to emulate and predict PDE-governed systems. My current research focuses on developing novel methods to learn stochastic or nonlocal closures for complex dynamical systems (e.g., multi-scale, multi-physics, chaotic) from limited data..

In general, my research interests lie in an interdisciplinary area of computational physics, applied mathematics and statistics.